不可切除局部晚期胰腺癌术中放疗临床效果的Meta分析

Meta-analysis of the clinical outcome of intraoperative radiotherapy in patients with unresectable locally advanced pancreatic cancer

目的：探讨术中放疗（IORT）治疗不可切除局部晚期胰腺癌的疗效及安全性。方法检索Cochrane图书馆、PubMed、EMbase、中国生物医学文献数据库（CBM）、中国知网（CNKI）、维普数据库、万方数据库文献，并手工检索重要文献的参考文献、会议论文，检索IORT治疗不可切除局部晚期胰腺癌的随机对照试验、队列研究、病例对照研究文献。按照纳入标准和排除标准选择文献，提取数据，利用RevMan5.3软件进行Meta分析。结果共检出1401篇文献，有7篇文献符合纳入标准，1篇为随机对照试验，4篇为队列研究，2篇为病例对照研究；IORT组共313例，对照组共362例，Meta分析结果显示IORT组总生存时间优于对照组（Z＝4.15，P＜0.0001，RR＝2，95%CI 1.05～2.94）。因局部控制率、疼痛缓解率、并发症情况均无完整数据，故仅进行描述性分析。结论 IORT可能提高不可切除局部晚期胰腺癌患者的总生存率，提高局部控制率，有效缓解疼痛，安全性较好，但仍需高质量的随机对照试验进一步证实。

Objective To explore the clinical efficacy and safety of intraoperative radiotherapy （IORT） for the patients with unresectable locally advanced pancreatic cancer. Methods Cochrane library, PubMed, EMbase, CBM, CNKI, VIP database and WANFANG database were used to retrieve, in addition to manual retrieval important literature references and conference papers. The randomised controlled trials, cohort study, controlled clinical trials of the IORT in unresectable locally advanced pancreatic cancer were searched. According to the inclusion criteria and exclusion criteria, after the extraction of literature and data, RevMan 5.3 software was used for meta-analysis. Results The total of 1 401 articles were retrieved, 7 met the inclusion criteria, including 1 for randomized controlled trials, 4 for cohort study, 2 for case-control study. 313 cases in the IORT group, 362 cases in the control group, and meta-analysis showed that IORT group was better than control group in overall survival （OS） （Z=4.15, P〈0.000 1, RR=2, 95%CI 1.05-2.94）. Because lack of complete data in local control rate, pain remission rate, complications, descriptive analysis was done, rather than meta-analysis. Conclusions IORT may improve the OS of patients with unresectable locally advanced pancreatic cancer and the local control rate, and effectively alleviate pain with favorale safety but it still needs high quality randomized controlled trials to confirm.